[R6RS] fixnum operators and overflow detection

[William D Clinger]

Mike quoting me:
> > If the editors agree with
> >
> > 1. adding the third option (fixnum operations that raise an
> > exception on overflow), and
> > 2. renaming the SRFI's fixnum operations to fixnum+, fixnum-,
> > fixnum-bitwise-and etc, and
> > 3. adding operations fx+, fx-, etc that raise an exception
> > on overflow,
> >
> > then
> >
> > 4. should I go ahead and perform that surgery on the SRFI
> > and on its reference implementation?
> 
> I think so.  Unless #1 and #3 are different, in which case I don't
> understand at least one of them.

#3 refers to the specific names of the operations mentioned
in #1.  Since there are other ways to choose the names, #3
is one of several ways to implement #1.

Anyway, I'll take this as assent to naming convention of #2
and #3, and will start to revise the arithmetic SRFI and its
implementation this weekend.  With luck, I might even finish
the revision.

Will